公开(公告)号：US10958901B2

公开(公告)日：2021-03-23

申请号：US16300691

申请日：2017-04-27

Applicant: B-COM

Inventor： Felix Henry ,  Joel Jung ,  Charlene Mouton

IPC: H04N19/107 ,  H04N19/176 ,  H04N19/11 ,  H04N19/119 ,  H04N19/184 ,  G06K9/62

Abstract: A method of decoding an encoded data stream representing at least one image subdivided into blocks. The method includes, for a current block of the image, and for each of at least two prediction modes available to predict the current block, obtaining a predictive block associated with the available mode; building a list having at least one of the available modes; for at least one mode included in the list, called a current mode, determining a distance measurement between the predictive block associated with the current mode and the other predictive blocks associated with the other available modes and obtained for the current block; modifying the list modified as a function of the determined measurement; decoding information identifying, for the current block, a prediction mode from the modified list from the encoded data stream; and rebuilding the current block from the predictive block associated with the identified prediction mode.

公开(公告)号：US10805638B2

公开(公告)日：2020-10-13

申请号：US16338878

申请日：2017-09-12

Applicant: B<>COM

Inventor： Felix Henry ,  Gordon Clare

IPC: H04N19/61 ,  H04N19/513

Abstract: A method for coding a digital image divided into blocks. The method includes, for a current block: processing the current block to provide a vector of transformed coefficients; selecting a sequence of M signs of coefficients to be predicted in the coefficient vector, M being an integer greater than or equal to 2; predicting values of the signs from a plurality of assumptions of combinations and a cost function; and for a sign of the sequence, coding an indicator representing a difference between its actual value and its predicted value. The predicting includes, for at least one second, current assumption: obtaining a pixel vector associated with the current assumption, by linear combination of at least one pixel vector associated with a first assumption and at least one pattern vector obtained from the vector of transformed coefficients; an calculating the cost function for the current assumption from the pixel vector obtained.

公开(公告)号：US10812814B2

公开(公告)日：2020-10-20

申请号：US16092962

申请日：2017-06-19

Applicant: B<>COM

Inventor： Joel Jung ,  Felix Henry ,  Charlene Mouton

IPC: H04N19/196 ,  H04N19/11 ,  H04N19/103 ,  H04N19/176 ,  H04N19/593 ,  H04N19/14

Abstract: A method for decoding a coded data stream representative of at least one image, the image being divided into blocks, a set of at least two prediction modes being available for predicting a current block. The method includes: obtaining at least two categories of prediction modes from the set; eliminating at least one category in accordance with at least one first predetermined elimination criterion; updating the categories not eliminated by deleting prediction modes that are common with the at least one eliminated category; and decoding, from the data stream, an information identifying for the current block a prediction mode among the modes of the updated categories.

公开(公告)号：US10798383B2

公开(公告)日：2020-10-06

申请号：US16320798

申请日：2017-07-12

Applicant: B<>COM

Inventor： Felix Henry ,  Gordon Clare

IPC: H04N19/13 ,  H04N19/136 ,  H04N19/184 ,  H04N19/132 ,  H04N19/85 ,  H04N19/91 ,  H04N19/176 ,  H04N19/70 ,  H04N19/895 ,  H04N19/40

Abstract: A method for decoding a stream having first and second portions portion and representing an image divided into blocks. The method includes a sequence test having a predetermined non-zero number of bits, including two at the power of the predetermined number (2N) iterations of the following substeps, for a current block: obtaining a sequence distinct from sequences already tested; decoding and reconstructing a version of the current block from the obtained sequence and from coded data in the first portion; evaluating a likelihood measurement associated with the reconstructed block; decoding information characteristic of a first sequence including the predetermined number of binary symbols from the second portion; selecting a sequence from the sequences tested, based on the likelihood measurements and the decoded information, the selected sequence being identified as the first sequence; and decoding and reconstructing the current block from the first sequence and the first portion.

公开(公告)号：US20190158841A1

公开(公告)日：2019-05-23

申请号：US16320798

申请日：2017-07-12

Applicant: B<>COM

Inventor： Felix Henry ,  Gordon Clare

IPC: H04N19/13 ,  H04N19/132 ,  H04N19/70 ,  H04N19/184 ,  H04N19/176

Abstract: A method for decoding a stream having first and second portions portion and representing an image divided into blocks. The method includes a sequence test having a predetermined non-zero number of bits, including two at the power of the predetermined number (2N) iterations of the following substeps, for a current block: obtaining a sequence distinct from sequences already tested; decoding and reconstructing a version of the current block from the obtained sequence and from coded data in the first portion; evaluating a likelihood measurement associated with the reconstructed block; decoding information characteristic of a first sequence including the predetermined number of binary symbols from the second portion; selecting a sequence from the sequences tested, based on the likelihood measurements and the decoded information, the selected sequence being identified as the first sequence; and decoding and reconstructing the current block from the first sequence and the first portion.

公开(公告)号：US10687082B2

公开(公告)日：2020-06-16

申请号：US16066966

申请日：2016-12-16

Applicant: B<>COM ,  ORANGE

Inventor： Felix Henry

IPC: H04N19/64 ,  H04N19/39 ,  H04N19/192 ,  H04N19/119 ,  H04N19/61 ,  H04N19/176 ,  H04N19/13 ,  H04N19/91 ,  H04N19/50 ,  H04N19/18 ,  H04N19/136 ,  H04N19/46

Abstract: A method for encoding a digital image divided into a plurality of blocks of pixels. The method includes, for a current block: processing the current block to provide a set of description elements; selecting a subset of description elements to be predicted; ordering the description elements of the subset into an ordered sequence; and encoding the elements of the sequence. Encoding includes scrolling the elements of the sequence and includes, for a current element: selecting a combination of predicted values of the description elements of the current sequence from a plurality of possible combinations based on a predetermined cost criterion and from the second element, values of description elements previously processed of the initial sequence; predicting the current element of the sequence by its value in the selected combination; and encoding an indicator representative of a difference between the actual value of the current element and its predicted value.

公开(公告)号：US20190208196A1

公开(公告)日：2019-07-04

申请号：US16300691

申请日：2017-04-27

Applicant: B-COM

Inventor： Felix Henry ,  Joel Jung ,  Charlene Mouton

IPC: H04N19/107 ,  H04N19/119 ,  H04N19/176 ,  H04N19/184 ,  G06K9/62

CPC classification number: H04N19/107 ,  G06K9/6215 ,  H04N19/11 ,  H04N19/119 ,  H04N19/176 ,  H04N19/184

Abstract: A method of decoding an encoded data stream representing at least one image subdivided into blocks. The method includes, for a current block of the image, and for each of at least two prediction modes available to predict the current block, obtaining a predictive block associated with the available mode; building a list having at least one of the available modes; for at least one mode included in the list, called a current mode, determining a distance measurement between the predictive block associated with the current mode and the other predictive blocks associated with the other available modes and obtained for the current block; modifying the list modified as a function of the determined measurement; decoding information identifying, for the current block, a prediction mode from the modified list from the encoded data stream; and rebuilding the current block from the predictive block associated with the identified prediction mode.

公开(公告)号：US20190174138A1

公开(公告)日：2019-06-06

申请号：US16092962

申请日：2017-06-19

Applicant: B<>COM

Inventor： Joel Jung ,  Felix Henry ,  Charlene Mouton

IPC: H04N19/196 ,  H04N19/103 ,  H04N19/14 ,  H04N19/593

CPC classification number: H04N19/197 ,  H04N19/103 ,  H04N19/11 ,  H04N19/14 ,  H04N19/176 ,  H04N19/593

Abstract: A method for decoding a coded data stream representative of at least one image, the image being divided into blocks, a set of at least two prediction modes being available for predicting a current block. The method includes: obtaining at least two categories of prediction modes from the set; eliminating at least one category in accordance with at least one first predetermined elimination criterion; updating the categories not eliminated by deleting prediction modes that are common with the at least one eliminated category; and decoding, from the data stream, an information identifying for the current block a prediction mode among the modes of the updated categories.

公开(公告)号：US11323715B2

公开(公告)日：2022-05-03

申请号：US16616252

申请日：2018-06-12

Applicant: FONDATION B-COM

Inventor： Felix Henry

IPC: H04B1/66 ,  H04N7/12 ,  H04N11/02 ,  H04N11/04 ,  H04N19/13 ,  H04N19/119 ,  H04N19/176 ,  H04N19/64 ,  H04L65/60

Abstract: A method for decoding an encoded data stream representative of at least one image, which is divided into blocks. The decoding method includes, for a current block: evaluating a plurality of value hypotheses of at least one description element of the current block, by calculating a likelihood measurement per hypothesis; calculating a disparity in the likelihood measurements obtained; determining at least one parameter of a decoder based on the calculated disparity; decoding, using the determined decoder, complementary information for identifying at least one of the hypotheses; and identifying at least one of the hypotheses using the decoded complementary information and obtaining a value of the at least one description element for the current block, from the at least one identified hypothesis.

公开(公告)号：US20190007706A1

公开(公告)日：2019-01-03

申请号：US16066966

申请日：2016-12-16

Applicant: B<>COM

Inventor： Felix Henry

IPC: H04N19/64 ,  H04N19/39 ,  H04N19/192 ,  H04N19/119 ,  H04N19/61 ,  H04N19/176

CPC classification number: H04N19/647 ,  H04N19/119 ,  H04N19/13 ,  H04N19/136 ,  H04N19/176 ,  H04N19/18 ,  H04N19/192 ,  H04N19/39 ,  H04N19/46 ,  H04N19/50 ,  H04N19/619 ,  H04N19/91

Abstract: A method for encoding a digital image divided into a plurality of blocks of pixels. The method includes, for a current block: processing the current block to provide a set of description elements; selecting a subset of description elements to be predicted; ordering the description elements of the subset into an ordered sequence; and encoding the elements of the sequence. Encoding includes scrolling the elements of the sequence and includes, for a current element: selecting a combination of predicted values of the description elements of the current sequence from a plurality of possible combinations based on a predetermined cost criterion and from the second element, values of description elements previously processed of the initial sequence; predicting the current element of the sequence by its value in the selected combination; and encoding an indicator representative of a difference between the actual value of the current element and its predicted value.